Skip to main content

Table 6 PCA dimensionality reduction results via 10-fold cross-validation for C. elegans

From: Comparative analysis and prediction of nucleosome positioning using integrative feature representation and machine learning algorithms

Method

Feature

Parameters

PCA%

ACC

Sn

Sp

MCC

AUC

SVM

FCGR

K = 1 + 2 + 4

0.88

0.8551

0.8960

0.8148

0.7130

0.9242

 

FCGR + DAC

K = 4, lag = 2

0.9

0.8562

0.8870

0.8259

0.7142

0.9245

 

FCGR + TAC

K = 4, lag = 2

0.93

0.8558

0.8824

0.8297

0.7132

0.9245

 

FCGR + DACC

K = 4, lag = 2

0.95

0.8265

0.9147

0.7397

0.6642

0.9057

 

FCGR + TACC

K = 4, lag = 2

0.95

0.8336

0.8079

0.8589

0.6682

0.9052

 

FCGR + PCPseDNC

K = 4, λ = 8, w = 0.5

0.85

0.8543

0.8913

0.8179

0.7112

0.9236

 

FCGR + PCPseTNC

K = 4, λ = 8, w = 0.5

0.95

0.8516

0.8929

0.8110

0.7064

0.9243

 

All features

 

0.95

0.8249

0.8029

0.8466

0.6513

0.8823

ELM

FCGR

K = 1 + 2 + 4

0.95

0.8535

0.8882

0.8194

0.7093

0.9193

 

FCGR + DAC

K = 4, lag = 2

0.88

0.8489

0.8742

0.8240

0.6990

0.9124

 

FCGR + TAC

K = 4, lag = 2

0.93

0.8500

0.8742

0.8263

0.7012

0.9157

 

FCGR + DACC

K = 4, lag = 2

0.9

0.8476

0.8703

0.8252

0.6962

0.9159

 

FCGR + TACC

K = 4, lag = 2

0.9

0.8537

0.8808

0.8271

0.7090

0.9183

 

FCGR + PCPseDNC

K = 4, λ = 8, w = 0.5

0.95

0.8466

0.8777

0.8160

0.6951

0.9158

 

FCGR + PCPseTNC

K = 4, λ = 8, w = 0.5

0.85

0.8452

0.8679

0.8229

0.6915

0.9160

 

All features

 

0.93

0.8505

0.8816

0.8198

0.7030

0.9183

XGBoost

FCGR

K = 1 + 2 + 4

0.90

0.8458

0.8870

0.8052

0.6946

0.9175

 

FCGR + DAC

K = 4, lag = 2

0.90

0.8526

0.8831

0.8225

0.7068

0.9234

 

FCGR + TAC

K = 4, lag = 2

0.95

0.8508

0.8738

0.8282

0.7028

0.9195

 

FCGR + DACC

K = 4, lag = 2

0.85

0.8396

0.8570

0.8225

0.6800

0.9147

 

FCGR + TACC

K = 4, lag = 2

0.95

0.8385

0.8621

0.8152

0.6782

0.9110

 

FCGR + PCPseDNC

K = 4, λ = 8, w = 0.5

0.93

0.8456

0.8808

0.8110

0.6934

0.9200

 

FCGR + PCPseTNC

K = 4, λ = 8, w = 0.5

0.90

0.8472

0.8835

0.8114

0.6967

0.9191

 

All features

 

0.95

0.8400

0.8613

0.8190

0.6812

0.9143

  1. “PCA%” means contributing rate of principal component
  2. Best values are in bold
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us